Masked Autoencoders for Microscopy are Scalable Learners of Cellular Biology

Featurizing microscopy images for use in biological research remains a significant challenge, especially for large-scale experiments spanning millions of images. This work explores the scaling properties of weakly supervised classifiers and self-supervised masked autoencoders (MAEs) when training wi...

Full description

Saved in:
Bibliographic Details
Main Authors: Kraus, Oren, Kenyon-Dean, Kian, Saberian, Saber, Fallah, Maryam, McLean, Peter, Leung, Jess, Sharma, Vasudev, Khan, Ayla, Balakrishnan, Jia, Celik, Safiye, Beaini, Dominique, Sypetkowski, Maciej, Cheng, Chi Vicky, Morse, Kristen, Makes, Maureen, Mabey, Ben, Earnshaw, Berton
Format: Conference Proceeding
Language:English
Subjects:
Biological system modeling
cell biology
cell morphology
Computer architecture
Fourier transforms
Microscopy
Pattern recognition
Self-supervised learning
SSL
Training
vision transformer
ViT
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Featurizing microscopy images for use in biological research remains a significant challenge, especially for large-scale experiments spanning millions of images. This work explores the scaling properties of weakly supervised classifiers and self-supervised masked autoencoders (MAEs) when training with increasingly larger model backbones and microscopy datasets. Our results show that ViT-based MAEs outperform weakly supervised classifiers on a variety of tasks, achieving as much as a 11.5% relative improvement when recalling known biological relationships curated from public databases. Additionally, we develop a new channel-agnostic MAE architecture (CA-MAE) that allows for inputting images of different numbers and orders of channels at inference time. We demonstrate that CA-MAEs effectively generalize by inferring and evaluating on a microscopy image dataset (JUMP-CP) generated under different experimental conditions with a different channel structure than our pretraining data (RPI-93M). Our findings motivate continued research into scaling self-supervised learning on microscopy data in order to create powerful foundation models of cellular biology that have the potential to catalyze advancements in drug discovery and beyond. Relevant code and select models released with this work can be found at: https://github.com/recursionpharma/maes_microscopy.
ISSN:2575-7075
DOI:10.1109/CVPR52733.2024.01117